Performance of PV Generation Feedback

Controllers: Power Factor versus Volt-VAR

Control Strategies

Ashish Agrawal

Thesis submitted to the faculty of the Virginia Polytechnic Institute and State University in partial

fulfillment of the requirements for the degree of

Master of Science

in

Electrical Engineering

Robert P. Broadwater, Chair

Virgilio A. Centeno

Jaime De La Reelopez

April 23rd 2015

Blacksburg, Virginia

Keywords: Constant Power Factor Controllers, Feedback Controllers, PV Generation, Voltage

Control, Volt-VAR Controllers

Performance of PV Generation Feedback Controllers: Power Factor versus Volt-VAR Control

Strategies

Ashish Agrawal

ABSTRACT

The variable nature of photovoltaic (PV) generation can cause voltage fluctuations in power

distribution systems. Feedback control can be used to minimize the voltage fluctuations. This

thesis presents the results obtained from comparing the control performance of two types of PV

generation feedback control, namely Volt-VAR control and constant power factor control. A three

minute PV generation transient is used to evaluate controller performance, where the transient

data used originated from one second measurements taken on an actual PV generator. Using the

three minute transient, a set of parametric studies are performed on both feedback control

strategies. The performance of the control strategies are compared as to voltage control on the

distribution feeder and also to the effect that the control may have on transmission system

voltage. In considering transmission system voltage, the reactive power drawn from the

substation during the transient is evaluated. Simulation results suggest that the choice of control

to be implemented should be based on both transmission and distribution system operational

concerns.

iii

Acknowledgements

I would like to thank all my well-wishers, including my friends and family, for their support

and encouragement during my studies. I would also like to extend my heartfelt gratitude to my

advisor and other committee members who guided me through the process of conducting

research.

iv

Table of Contents

ABSTRACT .........................................................................................................................................ii

Acknowledgements ......................................................................................................................... iii

Table of Contents ............................................................................................................................ iv

List of Figures ................................................................................................................................... v

List of Tables ................................................................................................................................... vi

1. Introduction ............................................................................................................................. 1

2. Constant Power Factor and Volt-VAR Controllers ................................................................... 4

2.1 Constant Power Factor Controller ................................................................................... 4

2.2 Volt-VAR Controller .......................................................................................................... 4

3. Case Study Description ............................................................................................................ 7

3.1 Circuit Description ............................................................................................................ 7

3.2 PV Data ............................................................................................................................. 8

3.3 Purpose of the Study ...................................................................................................... 10

4. Performance of Volt-VAR Controller ..................................................................................... 12

5. Performance of Constant Power Factor Controller ............................................................... 19

5.1 Switch Controllers .......................................................................................................... 22

6. Comparison of Controller Performances ............................................................................... 24

7. Conclusion ............................................................................................................................. 28

References .................................................................................................................................... 30

v

List of Figures

Figure 1.1 Nevada PV plant output on a sunny day and a partly cloudy day ................................. 2

Figure 2.1 Volt-VAR controller characteristic ................................................................................. 5

Figure 2.2 Volt-VAR control block diagram ..................................................................................... 5

Figure 3.1 Circuit under study (IEEE 123 bus system) .................................................................... 8

Figure 3.2 Variable PV generation over the analysis time period .................................................. 9

Figure 3.3 Fraction of load supplied by the PV generator .............................................................. 9

Figure 4.1 Qdeviation vs slope of Volt-VAR controller characteristic ............................................... 14

Figure 4.2 Qavg vs slope of Volt-VAR controller characteristic ...................................................... 15

Figure 4.3 ΔV vs slope of Volt-VAR controller characteristic ........................................................ 16

Figure 5.1 Qdeviation vs Power factor .............................................................................................. 20

Figure 5.2 Qavg vs Power factor ..................................................................................................... 20

Figure 5.3 ΔV vs Power factor ....................................................................................................... 21

Figure 6.1 Qdeviation for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging

mode) ............................................................................................................................................ 25

Figure 6.2 Qavg for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging

mode) ............................................................................................................................................ 25

Figure 6.3 ΔV for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging

mode) ............................................................................................................................................ 26

Figure 6.4 Qdeviation for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading

mode) ............................................................................................................................................ 26

Figure 6.5 Qavg for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading

mode) ............................................................................................................................................ 27

Figure 6.6 ΔV for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading

mode) ............................................................................................................................................ 27

vi

List of Tables

Table 3-1 Rating of the PV generator ............................................................................................. 7

Table 3-2 Circuit Loading in absence of PV generator .................................................................... 7

Table 3-3 PV generation variation over analysis time period ......................................................... 9

Table 3-4 Maximum rates of increase and decrease in PV generation ........................................ 10

Table 4-1 Qdeviation vs slope of volt-var controller characteristic with varying voltage set-points 17

Table 4-2 Qavg vs slope of volt-var controller characteristic with varying voltage set-points ...... 17

Table 4-3 ΔV vs slope of volt-var controller characteristic with varying voltage set-point.......... 18

Table 5-1 Performance of constant power factor controller in power factor lagging mode ....... 21

Table 5-2 Performance of constant power factor controller in power factor leading mode ...... 21

Table 5-3 Time periods when solar generation continuously increases or decreases ................. 23

Table 5-4 Performance of constant power factor controller when the solar generation

continuously increases or decreases ............................................................................................ 23

1

1. Introduction

The growing demand for electric power coupled with depleting conventional sources of power

has led to efforts to increase power generation worldwide using renewable sources of energy.

Solar energy is one of the fastest growing renewable energy sources. Solar photovoltaic (PV)

technologies have the potential to meet much of the electricity demand in the United States for

the next several decades and are projected to supply nearly 14% of US electricity demand by 2030

and 27% of the demand by 2050 [1]. One advantage of PV technology is generating electricity

without harming the environment.

Though PV generation provides an attractive alternative to fossil fuels, production of power

using PV presents a number of challenges due to the variable nature of the solar resource

received on the surface of the earth [2]. For instance, a PV plant output can vary by 90% over a

few seconds on a partly cloudy day, as shown in Figure 1.1 [3]. This rapid variation in the amount

of power generated by PV systems leads to voltage fluctuations, reverse power flows, voltage

and current unbalances, and power losses in both the transmission and distribution systems [4,

5]. One of the biggest challenges in the distribution system is to maintain the customer voltage

level in its allowed range [6].

Reactive power control strategies can be used to minimize the voltage fluctuations in

distribution systems. Traditionally constant power factor control or constant reactive power

control strategies have been used to regulate voltage levels at the PV generators themselves.

Reference [7] discusses a reactive power control strategy based on constant power factor control

and suggests that applying constant power factor control may overheat the PV inverter by drawing

excessive reactive power. Reference [8] analyses the constant power factor control of the power

conditioning system of a large-scale (~ 2 MW) PV system and suggests that it is possible to achieve

voltage regulation using the constant power factor control. A power factor control strategy based

on sensitivity analysis is discussed in [9], where it is proposed that voltage regulation can be

achieved with less reactive power consumption if a location dependent power factor value is

assigned to each PV inverter.

2

Figure 1.1 Nevada PV plant output on a sunny day (top) and a partly cloudy day (bottom)

(Source: North American Electric Reliability Corporation, “Accommodating high levels of variable generation.” April 2009. Used under fair use.)

Volt-VAR controllers are advanced controllers used to regulate voltage levels. The performance

of a Volt-VAR droop controller has been examined in [10, 11], where it is suggested that Volt-VAR

control can effectively minimize voltage fluctuations caused by the intermittent PV generation.

3

This thesis studies and evaluates the performance of two feedback controlled strategies,

namely Volt-VAR control and constant power factor control. The comparison involves voltage

control considerations for both the distribution and the transmission systems. Voltage control for

the transmission system will be measured by evaluating which controller draws the least reactive

power from the distribution substation. Voltage control for the distribution system will be

measured by evaluating which controller can best minimize customer level voltage deviations.

Section 2 of this thesis describes the two types of controllers. The circuit under study and the PV

data used in the study are described in section 3. Sections 4 and 5 discuss the performance of Volt-

VAR and constant power factor controllers respectively. A comparison of performances of the

controllers has been discussed in section 6. Section 7 discusses the conclusion and scope for future

study.

4

2. Constant Power Factor and Volt-VAR Controllers

This section describes the two types of controllers and discusses the parameters associated with

each type of controller.

2.1 Constant Power Factor Controller

Constant power factor controllers have been traditionally used in PV generation to regulate

voltage. As the name suggests the PV generation is operated in a constant power factor mode.

The power factor can be leading or lagging. In order to keep the PV generator at a constant power

factor, the ratio of the active power to the reactive power is maintained constant. This is achieved

by varying the reactive power output with changes in the active power output to maintain a

constant power factor.

In constant power factor control mode the terminal voltage of the generator is not monitored.

The controller monitors the power factor of the generator’s power output and accordingly

controls the VAR output of the PV generator to keep the power factor at the set point [12].

2.2 Volt-VAR Controller

A Volt-VAR controller regulates the voltage by supplying reactive power if the line voltage falls

below a specified voltage set-point and absorbing reactive power if the line voltage is above the

set-point. A typical characteristic used in Volt-VAR feedback control is illustrated in Figure 2.1 [13].

The terminal voltage of the PV generator is represented on the horizontal axis and the reactive

power supplied by the PV generator is represented on the vertical axis. The control block diagram

for the volt-var controller is shown in Figure 2.2 [13].The basic principle underlying this control is

that if the terminal voltage of the generator falls below the voltage set-point, the generator starts

supplying reactive power which leads to an increase in the terminal voltage. On the other hand, if

the terminal voltage exceeds the voltage set-point, the PV generator starts absorbing reactive

power which leads to a decrease in the terminal voltage.

5

Vset-point

VARLimit

VARLimit

Reactive Power

Voltage

Slope = GainSupplyReactivePower

AbsorbReactivePower

Figure 2.1 Volt-VAR controller characteristic

(Source: J. Jung, A. Onen, K. Russell, R.P. Broadwater, S. Steffel, and A. Dinkel, "Configurable, hierarchical, model-based, scheduling control with photovoltaic generators in power distribution circuits", Renewable Energy, vol. 76, pp. 318-329, 2015. Adapted under fair use.)

Electrical Grid

+ GainVset-point

-

V

VAR Limits

++

VAR

VAR

Figure 2.2 Volt-VAR control block diagram

(Source: J. Jung, A. Onen, K. Russell, R.P. Broadwater, S. Steffel, and A. Dinkel, "Configurable, hierarchical, model-based, scheduling control with photovoltaic generators in power distribution circuits", Renewable Energy, vol. 76, pp. 318-329, 2015. Adapted under fair use.)

6

A Volt-VAR controller is characterized by three parameters which are

Slope of the Volt-VAR controller characteristic

It should be noted that the slope of the volt-var characteristic is always negative. However,

we will use the absolute value of the slope in later sections for analysis purposes. The block

representing ‘Gain’ in Figure 2.2 refers to the slope of the volt-var controller. The slope of the

volt-var controller curve is a dimensionless quantity as it represents the change of percentage

available VARs with respect to the change in per unit voltage set-point.

The slope of the Volt-VAR controller characteristic is measured in terms of the angle made by

the Volt-VAR characteristic with the horizontal axis. The angle made by the Volt-VAR

characteristic with the horizontal axis is measured in the clockwise direction and lies between 0

and 90 degrees. This way, an increase in the angle made by the characteristic with the horizontal

axis represents an increase in the steepness of the characteristic.

Voltage set-point

The voltage set-point, as shown in Figure 2.1 is the voltage level when the controller changes

its operation from supplying reactive power to absorbing reactive power. The volt-var curve can

be shifted along the horizontal axis to change the voltage set-point.

Lower and upper limits on VAR generation

The amount of reactive power which can be absorbed or supplied by the volt-var controller

can be fixed at a certain percentage of the VAR rating of the PV generator. The block representing

‘VAR Limits’ in Figure 2.2 refers to the percentage available VAR of the volt-var controller.

Unlike the power factor controller, the Volt-VAR controller monitors the terminal voltage of

the PV generator and uses the voltage error to control the reactive power output of the generator

[12].

7

3. Case Study Description

This section describes the circuit used for the given study, the solar data used, and the

purpose of the study.

3.1 Circuit Description

Here the IEEE 123-bus system will be used to compare the performance of constant power

factor and Volt-VAR controllers. Figure 3.1 shows the circuit under study. Different points are

labeled in the figure to show the locations of the PV injections, the substation, feeder end points

(where voltage measurements are taken), and the recloser (where the reactive power flow from

the substation is measured).

It should be noted that three-phase voltages are measured at two feeder end points (as

denoted by the two sets of three phase voltages {V1, V2, V3} and {V4, V5, V6} in Figure 3.1), and

phase C voltage is measured at the third feeder end point (as denoted by V7 in Figure 3.1). The

reactive power flow is denoted by Qsub in Figure 3.1. Table 3-1 shows the rating of the PV

generator and Table 3-2 presents the circuit loading in the absence of the PV generator. It should

be noted that the circuit loading does not change during the simulations.

Table 3-1 Rating of the PV generator

kW 2300

kVA 2576

kVar Limit 2576

Volts 480

kVar @ Rated kW 1160.1

Table 3-2 Circuit Loading in absence of PV generator

Real Power (kW)

Reactive Power (kVAR)

Phase A 1455.21 580.71

Phase B 958.63 340.74

Phase C 1185.79 398.58

Total 3599.63 1320.03

8

Figure 3.1 Circuit under study (IEEE 123 bus system)

3.2 PV Data

The PV generation data used in this study is derived from one second PV generation data

obtained from an actual PV generator. The actual PV generation data was collected over a year,

and in the study here a small sample of the data was used, from 14:22:00 (t = 0) to 14:25:00 (t =

180 s) on a given day in April. Figure 3.2 shows the PV generation data used in the study here

over the period of analysis for the three phases.

Table 3-3 shows the percentage variation in PV generation over the three phases. Figure 3.3

presents the variation in the fraction of load supplied by the PV generator during the analysis

time period. The percentage of load supplied by the PV generator varies from 53.4% to 16.2% in

phase A, from 79.8% to 24.2% in phase B, and from 62.8% to 18.2% in phase C. Table 3-4 shows

the maximum rates of increase and decrease in PV generation and the time periods of those

changes.

9

Figure 3.2 Variable PV generation over the analysis time period

Figure 3.3 Fraction of load supplied by the PV generator

Table 3-3 PV generation variation over analysis time period

PV Data in (kW) Maximum

Generation

Minimum

Generation

Percentage

Variation

Phase A 776.996 236.101 69.61

Phase B 764.6 231.517 69.72

Phase C 744.974 215.506 71.07

Total 2286.57 683.124 70.12

10

Table 3-4 Maximum rates of increase and decrease in PV generation

PV generation increasing PV generation decreasing

Maximum rate of

change (kW/s)

Time of change Maximum rate of

change (kW/s)

Time of change

Phase A 29.053 14:22:35 –

14:22:48

27.499 14:22:54 –

14:23:07

Phase B 28.984 14:22:35 –

14:22:48

27.461 14:22:54 –

14:23:07

Phase C 28.615 14:22:35 –

14:22:48

27.314 14:22:54 –

14:23:07

3.3 Purpose of the Study

The aim is to evaluate controller performance relative to reactive power drawn from the

substation and relative to customer level voltage deviations. Reactive power effects at the

substation will be measured with two variables, the maximum and the average reactive power

drawn from the substation, as measured by Qmax and Qavg, respectively. The reactive power

drawn from the substation is denoted by Qsub. Qmax is defined as the maximum reactive power

drawn from the substation over the analysis time period. Qmax will be evaluated in terms of its

deviation from the reactive power drawn from the substation in the absence of the PV generator,

denoted by Qref. As can be seen from Table 3-2, the reactive power drawn from the substation in

the absence of PV generator is 1320 kVAR (i.e., Qref = 1320 kVAR). The deviation of Qmax from Qref

is denoted by Qdeviation. Customer level voltage deviation, as denoted by ΔV, is measured as the

sum of the difference between (Vi)max and (Vi)min for each of the seven feeder voltages during the

period of analysis. Qmax, Qavg, Qdeviation and ΔV are computed by:

Qmax = Max ((Qsub)t) (1)

Qdeviation = Qmax − Qref (2)

Qavg = 1

181∑ (Qsub)t

t=180 st=0 (3)

(Vi)max = Max ((Vi)t) (4)

(Vi)min = Min ((Vi)t) (5)

ΔVi = (Vi)max − (Vi)min (6)

ΔV = ∑ ΔVii=7i=1 (7)

11

where i varies from 1 to 7 and t varies from 0 to 180 s. In the above equations, (Qsub)t represents

the reactive power drawn from the substation at time ‘t’ and (Vi)t represents the ith voltage

measurement at time ‘t’, where 0 <= t <= 180.

Quasi-steady state power flow analysis is used to analyze the response of the system at each

second. In practice, the controller feedback generally operates faster than the sample rate of 1

second used here. In actual practice, the inverter controls may have a sample rate of few

hundred milliseconds, which is still large compared to the electrical time constants of the

distribution feeders.

12

4. Performance of Volt-VAR Controller

In order to study the performance of Volt-VAR feedback control during PV transients,

simulations were performed by varying the parameters of the Volt-VAR controller. As discussed

in section 2.2, a Volt-VAR controller is characterized by the slope of the Volt-VAR controller

characteristic, voltage set-point, and lower and upper limits on VAR generation. For the present

study, the lower and upper limits on the VAR generation were kept constant at 90% of the VAR

rating of the PV generator, and slope of the Volt-VAR controller characteristic and voltage set-

point were varied.

Here the slope of the Volt-VAR controller characteristic is measured in terms of the angle made

by the Volt-VAR characteristic with the horizontal axis. Here the angle made by the Volt-VAR

characteristic with the horizontal axis is measured in a clockwise direction and lies between 0

and 90 degrees. This way, an increase in the angle made by the characteristic with the horizontal

axis represents an increase in the steepness of the characteristic.

Figure 4.1 (a & b) shows the variation in deviation (Qdeviation) of maximum reactive power (Qmax)

drawn from the substation from the reference value (Qref) with the angle made by the Volt-VAR

controller characteristic with the horizontal axis and absolute value of the slope of the

characteristic respectively during the PV transient. The vertical axis of the graph in Figure 4.1

represents Qdeviation and the horizontal axis represents the angle of the Volt-VAR controller

characteristic slope in Figure 4.1 (a) and absolute value of the slope in Figure 4.1 (b). Figures 4.2

and 4.3 show the variation in the average reactive power drawn from the substation (Qavg) and

voltage deviation (ΔV), respectively, with respect to the angle made by the Volt-VAR controller

characteristic and slope of the Volt-VAR controller characteristic. The following can be observed

from figures 4.1 and 4.2:

Qdeviation remains close to zero until the angle made by the Volt-VAR controller

characteristic with the horizontal axis reaches close to 70 degrees, despite the fact that

there is significant PV generation during the transient. This is because below 70 degrees

the PV generator supplies needed reactive power, which keeps Qdeviation close to zero.

13

Qdeviation and Qavg remain insensitive to the variation in slope of the Volt-VAR controller

characteristic until the angle reaches 70 degrees. After that, Qdeviation and Qavg either

increase or decline depending upon the voltage set-point of the controller.

Qdeviation and Qavg are the lowest when the voltage set-point is 1.1 pu. This is because when

the terminal voltage of the generator is set to a higher value, the PV generator starts

supplying more reactive power, leading to a decrease in maximum and average reactive

power drawn from the substation.

From Figure 4.3 it can be seen that:

The voltage deviation (ΔV) is almost insensitive to the slope of the Volt-VAR controller

characteristic when the angle made by the characteristic with the horizontal axis varies

from 45 to 70 degrees for voltage set-points of 1 pu and 1.05 pu. After 70 degrees, ΔV dips

and then again peaks at different angles depending on the voltage set-point of the Volt-

VAR controller.

For the voltage set-point of 1.1 pu, ΔV constantly decreases when the angle varies from 45

degrees to around 72 degrees. After that, it starts increasing.

A voltage set point of 1.1 pu gives the lowest ΔV.

Based on figures 4.1, 4.2, and 4.3, a Volt-VAR controller with a voltage set-point of 1.1 pu and

which makes an angle of 65-70 degrees with the horizontal axis in the clockwise direction is

selected for comparison with the performance of the constant power factor controller. While

controller settings with an angle more than 70 degrees may provide slightly better performance,

the former is selected to provide a margin of error that can occur when controllers are

implemented in the field. The selected range of operation is circled in Figure 4.3(a).

Tables 4-1, 4-2, and 4-3 present the variation of Qdeviation, Qavg, and ΔV with the slope of the

Volt-VAR controller characteristic. From the tables, it can be seen that increasing the voltage set-

point beyond 1.1 pu increases the voltage deviations at the feeder ends. Also, the performance

of the volt-var controller does not change much when the voltage set-point is increased beyond

1.1 pu.

14

(a)

(b)

Figure 4.1 Qdeviation vs slope of Volt-VAR controller characteristic

15

(a)

(b)

Figure 4.2 Qavg vs slope of Volt-VAR controller characteristic

16

(a)

(b)

Figure 4.3 ΔV vs slope of Volt-VAR controller characteristic

17

Table 4-1 Qdeviation vs slope of volt-var controller characteristic with varying voltage set-points

Absolute

value of

slope of

volt-var

curve

Angle* made

by the vot-

var curve

with x-axis in

degrees

Qdeviation in kVAR with varying voltage set-point

Voltage

set-point

= 1 pu

Voltage

set-point

= 1.05 pu

Voltage

set-point

= 1.1 pu

Voltage

set-point

= 1.2 pu

Voltage

set-point

= 1.3 pu

1 45 20 -39 -107 -229 -344

1.8 61 52 -56 -179 -379 -583

3 71.57 101 -79 -273

6 80.54 203 -175 -476 -1089 -1089

9 83.67 287 -235 -708 -1089 -1089

13 85.61 370 -298 -917 -1089 -1089

18 86.83 461 567 -1089 -1089 -1089

30 88.1 562 1634 -1089 -1089 -1089

*The angle made by the volt-var characteristic curve is measured from the horizontal axis (x-

axis) in the clockwise direction.

Table 4-2 Qavg vs slope of volt-var controller characteristic with varying voltage set-points

Absolute

value of

slope of

volt-var

curve

Angle* made

by the vot-

var curve

with x-axis in

degrees

Qavg in kVAR with varying voltage set-point in per unit

Voltage

set-point

= 1 pu

Voltage

set-point

= 1.05 pu

Voltage

set-point

= 1.1 pu

Voltage

set-point

= 1.2 pu

Voltage

set-point

= 1.3 pu

1 45 442 409 375 306 236

1.8 61 453 395 335 213 97

3 71.57 467 375 276

6 80.54 496 333 158 -183 -190

9 83.67 519 278 43 -190 -190

13 85.61 542 241 -80 -190 -190

18 86.83 479 372 -162 -190 -190

30 88.1 -2 779 -190 -190 -190

*The angle made by the volt-var characteristic curve is measured from the horizontal axis (x-

axis) in the clockwise direction.

18

Table 4-3 ΔV vs slope of volt-var controller characteristic with varying voltage set-point

Absolute

value of

slope of

volt-var

curve

Angle* made

by the vot-

var curve

with x-axis in

degrees

ΔV in volt with varying voltage set-point in per unit

Voltage

set-point

= 1 pu

Voltage

set-point

= 1.05 pu

Voltage

set-point

= 1.1 pu

Voltage

set-point

= 1.2 pu

Voltage

set-point

= 1.3 pu

1 45 5.55 5.23 4.47 2.53 4.21

1.8 61 5.47 4.57 2.88 4.2 3.74

3 71.57 5.49 4.39 2.24

6 80.54 5.47 2.73 3.65 4.54 4.63

9 83.67 4.54 4.47 4.47 4.63 4.63

13 85.61 4.55 4.47 4.47 4.63 4.63

18 86.83 5.55 4.47 4.47 4.63 4.63

30 88.1 11.28 4.47 4.63 4.63 4.63

*The angle made by the volt-var characteristic curve is measured from the horizontal axis (x-

axis) in the clockwise direction.

19

5. Performance of Constant Power Factor Controller

In order to study the performance of the constant power factor controller, simulations were

performed at different power factors values for both leading and lagging modes.

Figure 5.1 shows the variation of Qdeviation with varying power factor for both leading and

lagging modes. Figures 5.2 and 5.3 present the variation in average reactive power drawn from

the substation (Qavg) and voltage deviation (ΔV), respectively, with variation in power factor of

the PV generator.

The following observations can be made from figures 5.1, 5.2, and 5.3 for the power factor

lagging mode:

Qdeviation increases with increasing power factor, but the Qdeviation values are always less

than zero (Figure 5.1). This implies that the maximum reactive power drawn from the

substation when the PV generator is run in power factor lagging mode is always less than

the reactive power drawn from the substation in the absence of the PV generator (Qref).

The average reactive power drawn from the substation (Qavg) increases with increasing

power factor values (Figure 5.2).

Voltage deviation (ΔV) values do not change significantly with power factor and lie

between 5 and 6 V for the power factor values between 0.9 and 1 (Figure 5.3).

The following can be seen from figures 5.1, 5.2, and 5.3 for the power factor leading mode:

Qdeviation decreases with increasing power factor magnitude, where the Qdeviation values are

always greater than zero (Figure 5.1). This implies that the maximum reactive power

drawn from the substation when the PV generator is run in power factor leading mode is

always greater than the reactive power drawn from the substation in the absence of the

PV generator (Qref).

The average reactive power drawn from the substation (Qavg) decreases when the

magnitude of power factor increases from 0.9 to 0.98 (Figure 5.2).

For power factors between 0.92 and 0.98, voltage deviation (ΔV) values are low and lie

between 1.56 and 2.38 V (Figure 5.3).

20

Figure 5.1 Qdeviation vs Power factor

Figure 5.2 Qavg vs Power factor

21

Figure 5.3 ΔV vs Power factor

Tables 5-1 and 5-2 present the performances of constant power factor controller in

lagging and leading modes respectively.

Table 5-1 Performance of constant power factor controller in power factor lagging mode

Power Factor Qmax (kVAR) Qdeviation (kVAR) Qavg (kVAR) ΔV (volt)

.9 954 -366 214 5.47

.92 994 -326 239 5.47

.94 1035 -285 267 5.79

.96 1084 -236 298 5.90

.98 1146 -174 337 5.55

1 1302 -18 428 5.46

Table 5-2 Performance of constant power factor controller in power factor leading mode

Power Factor Qmax (kVAR) Qdeviation (kVAR) Qavg (kVAR) ΔV (volt)

-.9 2464 1144 650 5.51

-.92 2324 1004 623 1.90

-.94 2168 848 594 1.60

-.96 1993 673 560 1.56

-.98 1781 461 520 2.38

22

From figures 5.1, 5.2, and 5.3, it can be seen that operating the PV generator at a lagging

power factor minimizes the reactive power drawn from the substation, but it increases the

voltage deviations. On the other hand, operating the PV generator in leading power factor mode

minimizes the voltage deviation at the feeder ends at the cost of drawing more reactive from the

substation. This finding was used to develop the idea for a new type of controller, called ‘switch

controller’. The basic idea behind switch controller and the evaluation of its performance is

described in the following subsection.

5.1 Switch Controllers

As discussed earlier, operating the PV generator at a lagging power factor minimizes the

reactive power drawn from the substation at the expense of increasing the voltage deviations at

the customer end. On the other hand, operating the PV generator in leading power factor mode

minimizes the voltage deviations at the feeder end at the cost of drawing more reactive from the

substation. The idea of switch controllers was developed by switching the power factor from

leading to lagging when the solar generation started to increase continuously and from lagging

to leading when the generation started to decrease continuously. For this, three time periods

were identified when the solar generation was either continuously increasing or decreasing.

These time periods are shown in Table 5-3.

As can be seen from the Table 5-3, the solar generation continuously increased during T2

and it continuously decreased during T1 and T3. The performances of constant power factor

controller at a power factor of 0.94 both in leading and lagging modes were evaluated for the

times periods T1, T2, T3. Table 5-4 shows the performances of constant power factor controllers

during this these time periods. It can be seen that the performances of power factor controllers

are similar during the three time periods irrespective of whether the generation is increasing or

decreasing. Also, the performances of the controller in leading or lagging mode during the three

time periods (T1, T2, T3) were not different from the performance of the controller during the

entire duration of analysis.

23

Table 5-3 Time periods when solar generation continuously increases or decreases

Time Period Solar Geneation

T1: 14:22:04 – 14:22:29 Decreasing

T2: 14:22:35 – 14:22:48 Increasing

T3: 14:22:54 – 14:23:07 Decreasing

Table 5-4 Performance of constant power factor controller when the solar generation continuously increases or decreases

Power Factor Duration of Analysis Qdeviation (kVAR) ΔV (volt)

0.94 T1 -340 5.21

0.94 T2 -341 4.75

0.94 T3 -426 4.79

0.94 Full -285 5.79

-0.94 T1 775 1.26

-0.94 T2 770 1.30

-0.94 T3 815 1.19

-0.94 Full 848 1.60

This suggests that if we switch the controller from leading to lagging during the period when

solar generation increases continuously, we will be able to limit the reactive power flowing from

the substation during the period when the controllers are switched but won't be able to regulate

the voltage at the customer end. On the other hand, switching the controller from lagging to

leading when the solar generation decreases continuously will control voltage deviations at the

customer ends but will draw large amounts of reactive power from the substation.

24

6. Comparison of Controller Performances

This section compares the performances of the two feedback control strategies. To compare

the performance of the Volt-VAR control with the constant power factor control, two different

constant power factor controllers will be used. In comparing the reactive power drawn from the

substation, the lagging power factor control is used, and in comparing the voltage deviation the

leading power factor control is used.

As discussed in Section 4, a Volt-VAR controller with a voltage set-point of 1.1 pu, which makes

an angle of 65-70 degrees with the horizontal axis in clockwise direction, is selected for

comparison with the constant power factor control.

Figures 6.1 and 6.2 compare the performance of Volt-VAR and constant power factor control

in lagging mode as to Qdeviation and Qavg. The following observations can be made from these

figures:

For the selected range of operation of the Volt-VAR controller, Qdeviation is higher for the

Volt-VAR controller when the power factor is less than 0.96. Qdeviation is lower for the Volt-

VAR controller when the power factor is more than 0.96 (Figure 6.1).

Qavg follows the same trend as Qdeviation (Figure 6.2).

It should be noted that the voltage deviation values at the feeder ends (ΔV) are always higher

for the power factor controller in lagging mode as compared to that for the Volt-VAR controller

with a voltage set-point of 1.1 pu (Figure 6.3).

Figures 6.4, 6.5, and 6.6 compare the performances of the Volt-VAR controller and the power

factor controller in leading mode. It should also be noted that the Qdeviation and Qavg for the Volt-

VAR controller are always lower than those obtained using a constant power factor controller in

leading mode (Figures 6.4 & 6.5).

Figure 6.6 compares the performances of the Volt-VAR controller and constant power factor

controller in leading mode on the basis of ΔV. It can be seen from Figure 6.3 that for the selected

range of operation of the Volt-VAR controller, ΔV is higher than that obtained from a power factor

controller operating at a leading power factor of 0.92 or higher.

25

Figure 6.1 Qdeviation for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging mode)

Figure 6.2 Qavg for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging mode)

26

Figure 6.3 ΔV for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (lagging mode)

Figure 6.4 Qdeviation for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading mode)

27

Figure 6.5 Qavg for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading mode)

Figure 6.6 ΔV for Volt-VAR controller (Vset-point = 1.1 pu) & power factor controller (leading mode)

28

7. Conclusion

Parametric studies were run for both Volt-VAR and power factor controllers, and the

performance of the controllers were compared as to transmission system reactive power

demands and customer level voltage deviations. In the control comparisons it was found that the

constant power factor controller operating in lagging mode minimizes the reactive power drawn

from the substation at the expense of increased customer level voltage deviations. On the other

hand, the power factor controller operating in leading mode minimizes the customer level

voltage deviations at the expense of increasing the reactive power drawn from the substation.

Simulation results show that the voltage deviation is lower for the Volt-VAR controller in the

selected range of operation as compared to that obtained for the constant power factor

controller operating in the lagging mode. On the other hand, maximum reactive power and

average reactive power drawn from the substation are lower for the Volt-VAR controller in the

selected range of operation as compared to those obtained for constant power factor controller

operating in leading mode. However, in the study here the constant power factor controller

operating in the leading mode provides better voltage control than the Volt-VAR controller.

The choice of controller to be implemented should be based on both transmission and

distribution system concerns. If the major aim is to regulate the transmission system voltage, a

controller which draws the least amount of reactive power from the substation should be used.

Constant power factor controllers operating in lagging mode are suitable for this. On the other

hand, if the major aim is to reduce voltage fluctuations in the distribution system, a controller

which best regulates the voltage at the customer end should be used. In the study here, constant

power factor controllers operating in leading mode are best suited for this. If the aim is to take

into account both concerns, then the Volt-VAR controller could be a good compromise.

The current study investigated and compared the performances of Volt-VAR and constant

power factor controllers in case of one PV generator present in the system. Future work could

evaluate the performance of controllers when multiple PV generators are present in the system.

Analysis can be performed to study how the performance changes for a given controller when

different PV generators in the system are controlled by different types of controllers.

29

Performance of a new type of controller which regulates the power factor of the PV generator

based on its terminal voltage can also be evaluated.

30

References

[1] United States Department of Energy, SunShot Vison Study, Report DOE/GO-102012-3037, Feb. 2012. Available at: < http://www1.eere.energy.gov/solar/sunshot/vision_study.html>.

[2] G K Singh, “Solar power generation by PV (photovoltaic) technology: A review”, Energy, vol. 53, pp. 1-13, 2013.

[3] North American Electric Reliability Corporation, “Accommodating high levels of variable generation.” April 2009.

[4] F. Katiraei, and J. R. Aguero, "Solar PV integration challenges," Power and Energy Magazine, IEEE , vol.9, no.3, pp.62,71, May-June 2011.

[5] W. Zhang, M. Baran, A. De, and S. Bhattacharya, "Fast volt-VAR control on PV dominated distribution systems," 2014 IEEE PES T&D Conference and Exposition, vol., no., pp.1,5, 14-17 April 2014.

[6] W. Shang, and M.A. Redfern, "A new control scheme for a distributed generator providing network voltage support," 2011 International Conference on Advanced Power System Automation and Protection (APAP), vol.3, no., pp.2453,2457, 16-20 Oct. 2011.

[7] O. Phochai, W. Ongsakul, and J. Mitra, "Voltage control strategies for grid-connected solar PV systems," 2014 International Conference and Utility Exhibition on Green Energy for Sustainable Development (ICUE), vol., no., pp.1,7, 19-21 March 2014.

[8] K. Kouno, M. Hirose, T. Nagase, and W. Hashimoto, W.; Hojo, M.; Ohnishi, T., "Effectiveness of constant power factor control of mega-solar system for voltage regulation with long distribution line," 2013 IEEE 10th International Conference on Power Electronics and Drive Systems (PEDS), vol., no., pp.317,322, 22-25 April 2013.

[9] E. Demirok, P. Casado Gonzalez, K.H.B. Frederiksen, D. Sera, P. Rodriguez, and R. Teodorescu, "Local reactive power control methods for overvoltage prevention of distributed solar inverters in low-voltage grids," IEEE Journal of Photovoltaics, vol.1, no.2, pp.174,182, Oct. 2011.

[10] J.W. Smith, W. Sunderman, R. Dugan, and B. Seal, "Smart inverter volt/var control functions for high penetration of PV on distribution systems," 2011 IEEE/PES Power Systems Conference and Exposition (PSCE), vol., no., pp.1,6, 20-23 March 2011.

[11] H. Alatrash, R.A. Amarin, and C. Lam, "Enabling large-Scale PV integration into the Grid," 2012 IEEE Green Technologies Conference, vol., no., pp.1,6, 19-20 April 2012.

[12] W. Shang, S. Zheng, L. Li, and M. Redfern, "A new volt/VAR control for distributed generation," 2013 48th International Universities' Power Engineering Conference (UPEC), vol., no., pp.1,5, 2-5 Sept. 2013.

[13] J. Jung, A. Onen, K. Russell, R.P. Broadwater, S. Steffel, and A. Dinkel, "Configurable, hierarchical, model-based, scheduling control with photovoltaic generators in power distribution circuits", Renewable Energy, vol. 76, pp. 318-329, 2015.